| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology and Intensive Care, Hôpital J. Bernard, Poitiers, France; Department of Anesthesiology and Intensive Care, Hôpital R. Poincaré, Garches, France
Address correspondence and reprint requests to Myriam Tirault, MD, Département d'Anesthésie-Réanimation, C.H.U. La Milétrie, BP 577, 86021 Poitiers Cedex, France. Address e-mail to m.tirault{at}chu-poitiers.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Animal experiments (1) and volunteer studies (2) have reported that large-dose opioid administration may result in the phenomenon of "acute opioid tolerance." Two clinical studies (3,4) have demonstrated acute opioid tolerance after inhaled anesthesia, but this has not been shown during total IV anesthesia. Acute tolerance was manifested by increased postoperative pain, larger morphine requirement, and hyperalgesia. Others failed to confirm acute opioid tolerance (5–7), probably because conditions for its development were not present: painful surgery, large opioid dosage, use of short-acting opioids, and extended exposure (at least 90 min).
Acute opioid tolerance reflects a central sensitization of the neuronal systems involved in the processing of nociceptive information. There is some evidence suggesting that glutamate, a spinal excitatory amino acid, plays a critical role in nociceptive neurotransmission (8,9). N-methyl-d-aspartate (NMDA) receptors are one of the major subtypes of glutamate receptors in the spinal cord. Acute tolerance to opioids may result in an enhancement in glutamatergic release, which in turn reduces the analgesic effects of opioids. More specifically, remifentanil may directly activate the NMDA receptors (10). An alternative to avoid acute opioid tolerance may be to coadminister drugs that prevent NMDA activation.
Ketamine, an NMDA receptor antagonist, blocks development of morphine tolerance (11–13). In a recent study, nefopam hydrochloride was as effective as ketamine in producing a morphine-sparing effect in resistant patients (14). Nefopam is a potent analgesic with a central mechanism of action (15). Its antinociceptive effect involves the inhibition of monoamine reuptake (16). Recently, Verleye et al. (17) suggested that nefopam may also modulate glutamatergic neurotransmission. Thus, nefopam should possibly reduce postoperative morphine overconsumption.
The purpose of this study was twofold: to confirm whether remifentanil caused morphine overconsumption after IV anesthesia and to investigate whether nefopam would reduce this. To test these hypotheses, postoperative morphine consumption was compared between two groups of patients undergoing major abdominal surgery and receiving either small- or large-dose remifentanil combined with propofol. The effects of nefopam on postoperative morphine requirement were then studied.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study exclusion criteria were ASA physical status IV–V, history of chronic pain, use of analgesics or opioids within 12 h before surgery, drug or alcohol addiction, psychiatric disorder, body mass index (BMI) >35, pregnancy, breast-feeding, presence of hepatic, kidney or pulmonary dysfunction, contraindications to any drug used in the study, inability to understand self-administration of opioids, and participation in another research project in the previous 30 days.
The evening before surgery, patients were instructed on the use of a 100-mm visual analog scale (VAS) with 0 mm identifying no pain and 100 mm being defined as the worst imaginable pain. They were also instructed on the use of the patient-controlled-analgesia (PCA) device (Fresenius Vial, Brezins, France). Patients were premedicated with their usual medicaments if necessary and with hydroxyzine 1 mg/kg orally the night before and 1 h before surgery.
The IV anesthetic technique was standardized and administrated by the same anesthesiologist. The Ethics Committee insisted for ethical reasons that the study had to be conducted in two separate parts.
Part I: The Effect of Large-Dose Remifentanil During Propofol Anesthesia on Postoperative Morphine Consumption
In the operating room, patients were connected to routine physiological monitors: pulse oximetry, electrocardiogram, and noninvasive arterial blood pressure. Baseline values, defined as the mean of the two lowest measurements just before induction of anesthesia, were recorded. Measurements were repeated at 5-min intervals thereafter. A cannula was inserted into a vein of the forearm for administration of anesthetic drugs. Lactated Ringer's solution or saline was infused at 10 mL · kg–1 · h–1. Patients were administered oxygen for 3 min by mask with 100% oxygen and anesthesia was started.
Anesthesia was provided by target-controlled infusions of remifentanil and propofol using the pharmacokinetic datasets published by Minto et al. (18) and Gepts et al. (19,20) respectively. Remifentanil was infused in target effect-site mode using STANPUMP® software (Dr. Steven Shafer, Palo Alto, CA) and a Pilote Anesthesie® pump (Becton Dickinson, Brezins, France) driven by a personal computer. Propofol was given in target plasma concentration mode using Diprifusor® (Master TCI, Fresenius Vial).
For induction of anesthesia, the target propofol concentration was programmed at 6–8 µg/mL and the target remifentanil concentration at 4 ng/mL. After loss of consciousness, patients received a bolus of atracurium 0.5 mg/kg. The trachea was then intubated and the lungs of the patients mechanically ventilated to obtain normocapnia with 50% oxygen in nitrous oxide.
The first 30 patients were randomly allocated to 2 groups of 15 by an anesthesiologist who did not participate in patient care. Allocation was maintained in opaque envelopes. After induction of anesthesia, one of these was opened by the anesthesiologist involved in patient care. This practitioner did not participate in patient evaluation during the postoperative period. Furthermore, patients, surgeons, and all nurses were blinded as to the intraoperative management.
Target remifentanil concentration in the two groups of patients was not changed during surgery. In the small-dose group (Group S), a remifentanil effect-site concentration of 3 ng/mL was targeted, and in the large-dose group (Group L) it was 8 ng/mL. The target plasma propofol concentration was adjusted between 3 µg/mL and 8 µg/mL, according to the patients' autonomic responses.
Hemodynamic variables (heart rate, systolic arterial blood pressure) where maintained within 20% of baseline. When anesthesia was insufficient (high systolic arterial blood pressure or tachycardia during at least 1 min, movement, coughing, tearing, or sweating), the propofol concentration was increased by 1 µg/mL. When mean arterial blood pressure was too low, it was decreased by 1 µg/mL and the patient received additional IV fluids or ephedrine if necessary. If heart rate decreased to 45 per min, IV atropine 0.5 mg was administrated. Atracurium infusion was titrated to maintain two twitches in response to a supramaximal train-of-four stimulation applied on the facial nerve. Hemodynamic variables, pulse oximetry, and end-tidal carbon dioxide pressure, train-of-four count, and temperature were recorded every 10 min. All patients were laid under a warming blanket to maintain core temperature at approximately 36°C.
Thirty minutes before the anticipated end of surgery, all patients received an injection of IV morphine 0.15 mg/kg. Propofol and remifentanil target concentrations were set to zero at the final skin sutures.
After arrival in the postanesthesia care unit (PACU), residual neuromuscular blockade was reversed with neostigmine (40 µg/kg) and atropine (15 µg/kg). The endotracheal tube was removed when the patient recovered consciousness (eye opening or purposeful movement) and spontaneous respiratory rate exceeded 12 breaths/min. Pain intensity was assessed using a VAS score. Pain was controlled only by titration of IV morphine administered by nurses who were blinded as to the group assignment. When the VAS score reached 30, IV morphine was given every 5 min in 3-mg increments (2 mg in patients aged 
65 yr) until pain relief, defined as a VAS score of 30 or less. Clinical monitoring included respiratory rate, oxygen saturation, sedation according to the Ramsay score, arterial blood pressure, and heart rate. Morphine titration was stopped if the patient had a respiratory rate less than 12 breaths/min, oxygen saturation less than 95%, or experienced a serious adverse event related to morphine administration (rash, vomiting, pruritus). When the VAS score was 30 or less, morphine titration was stopped, and analgesia continued with PCA delivering morphine, 1 mg as an IV bolus, with a lockout interval of 8 min and no background infusion or limits.
Physiological variables were recorded at 15-min intervals during the first 30 min and then every 30 min. After 4 h in the PACU, patients returned to their surgical ward. Noninvasive arterial blood pressure, heart rate, respiratory rate, morphine consumption, VAS score, Ramsay score, nausea, vomiting, and other complications (pruritus, hallucinations, tachycardia, sweating, shivering) were recorded every 4 h for a further 20 h. Nausea and vomiting were treated with droperidol 1 mg IV.
Duration of anesthesia (from the beginning of the studied drugs injection to the end of drug delivery), remifentanil and propofol total doses, and time of the intraoperative morphine bolus, were recorded manually. In the PACU, tracheal extubation delay was defined as the time from the end of anesthetic infusion to the endotracheal tube removal. The time for the first morphine demand was defined as the duration from extubation to the first request for analgesic medication. The morphine titration requirement (expressed as mg/kg) was the amount of morphine needed to obtain pain relief (VAS score 
30) during IV morphine titration and did not consider the intraoperative morphine bolus. The PCA morphine consumption was the amount of morphine administrated by PCA device throughout the first 24 postoperative hours, excluding titration.
Part II: The Effect of Nefopam on Postoperative Morphine Consumption After Large-Dose Remifentanil and Propofol-Based Anesthesia
Thirty additional patients were randomly allocated in 2 groups of 15. The anesthetic protocol was the same as in Part I, but patients also received nefopam 20 mg 30 min before the end of surgery with the intraoperative morphine bolus.
Because the primary outcome of the study was the amount of morphine given by IV titration in the PACU, the number of patients was calculated according to this variable. Reported postoperative morphine titration varies widely. Our experience has indicated that morphine titration after major abdominal surgery in patients receiving large-dose remifentanil was approximately 20 mg (21). The sample size was calculated to detect a difference in morphine titration requirement of at least 50% between the groups receiving the smallest and the largest doses of remifentanil: 15 patients per group would give a power of 0.8 at an 
level of 0.05.
Quantitative variables (age, weight, height, time delays, duration of anesthesia, intraoperative doses of remifentanil and propofol, morphine given by titration, PCA morphine consumption over 24 h) were reported as median ± range. The Mann-Whitney U-test was used to compare the two groups. The fractions of patients not starting morphine titration in PACU were evaluated with survival curves and were compared with the Kaplan-Meier log-rank test. The 
2 test was used for categorical variables (gender, ASA physical status, incidence of emesis). Repeated-measures analyses of variance were used for continuous variables (VAS score). Statistical analysis was performed using Statview® software 5.0 (SAS Institute, Cary, NC). A value of P 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Part I: The Effect of Large-Dose Remifentanil During Propofol Anesthesia on Postoperative Morphine Consumption
Thirty patients were enrolled for the first set of experiments, 15 in Group S and 15 in Group L. Two patients in Group S were withdrawn from the study, one because morphine titration was started before tracheal extubation and one because intraoperative anesthesia was insufficient despite a remifentanil target concentration of 12 ng/mL and large-dose propofol, which resulted in a delayed recovery (3 h after the end of surgery). Patient characteristics and duration of anesthesia were similar in the two groups (Table 1).
|
The median rate of remifentanil infusion calculated by STANPUMP® software was 0.08 µg · kg–1 · min–1 in Group S and 0.24 µg · kg–1 · min–1 in Group L (P < 0.0001). Propofol consumption was significantly reduced by 30% in Group L compared with Group S: 0.14 mg · kg–1 · min–1 and 0.2 mg · kg–1 · min–1 respectively (P < 0.0008). Duration of anesthesia and postoperative care data are summarized in Table 2.
|
Intensity of pain measured with VAS was similar in the 2 groups during the 4 h in PACU and during the next 20 h in the surgical ward (data not shown). All patients in both groups required titration before PCA initiation. Survival curve analysis of the first morphine administration (Fig. 1A) indicate that patients in Group L requested morphine by titration significantly earlier (10 [1–63] min) than did those in Group S (37 [5–90] min; P < 0.002). The dose of morphine given by titration in PACU was significantly larger in Group L than in Group S (Table 2 and Fig. 2). PCA morphine requirements did not differ between the two groups at the end of the 24 h postoperatively.
|
|
Hemodynamic variables remained constant throughout the study. One patient in Group L experienced a deep bradycardia (35/min) for 1 min without low arterial blood pressure when the remifentanil target concentration was increased dramatically to 8 ng/mL before skin incision. The patient recovered rapidly with administration of atropine 0.5 mg.
There were no differences between groups in respiratory rate, Spo2, and sedation scores. Nausea and vomiting were the most prevalent adverse events.
Part II: The Effect of Nefopam on postoperative Morphine Consumption After Large-Dose Remifentanil and Propofol-Based Anesthesia
Patient characteristics and duration of anesthesia were similar in Groups small dose with nefopam (Sn) and large dose with nefopam (Ln) and are summarized in Table 1. One patient in Group Sn was withdrawn from the study because he received the nefopam injection just after induction of anesthesia instead of 30 min before skin closure.
Remifentanil and propofol consumption differed significantly: 0.08 µg · kg–1 · min–1 remifentanil and 0.19 mg · kg–1 · min–1 propofol in Group Sn; 0.2 µg · kg–1 · min–1 remifentanil and 0.13 mg · kg–1 · min–1 propofol in Group Ln (P < 0.0001 and P < 0.002 respectively). VAS scores were similar throughout all the 24 postoperative hours (data not shown). Four patients in Group Sn and one in Group Ln did not demand any analgesic medication immediately after tracheal extubation. One patient in Group Sn did not demand any morphine by PCA during the first 24 h postoperatively. The time to the first morphine request in PACU did not differ during Part II, as presented on Fig. 1B (17 [1–57] min in Group Sn; 20 [6–70] min in Group Ln). The dose of morphine given by titration was not significantly increased in group Ln compared with group Sn (Table 2 and Fig. 2). PCA morphine requirements did not differ between the two groups. Intraoperative administration of nefopam did not produce any tachycardia or profuse sweating. Neither hemodynamic or respiratory disorders nor excessive sedation or nausea were noted during Part II.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Acute opioid tolerance induced by large-dose opioid has previously been demonstrated during inhaled anesthesia (3,4) but not during total IV anesthesia. Part I was conducted to verify this effect with remifentanil during IV anesthesia. Part II was designed to test the hypothesis that nefopam would reduce the amount of morphine required to produce adequate analgesia after large doses of remifentanil. Thus, Part II was initiated once the results of Part I confirmed the existence of acute opioid tolerance during propofol-based anesthesia.
This study was not designed in a double-blind fashion during surgery for several reasons. First, when using STANPUMP® software, the infusion rate of remifentanil during the procedure corresponding to the group assignment appeared, and it was not possible to hide the computer screen. Second, safety was enhanced when the physician in charge of the patient knew which dose of remifentanil was infused. However, as the medical staff involved in postoperative pain evaluation was blinded, our postoperative results, specifically the morphine titration in PACU, were not affected by the lack of intraoperative blinding.
Our anesthetic protocol included conditions for development of acute opioid tolerance: abdominal laparotomy is one of the most painful surgeries and duration of anesthesia was more than 90 minutes; therefore, remifentanil was chosen as the intraoperative opioid because it is a potent and short-acting drug that directly activates NMDA receptors (10). A bolus of morphine 0.15 mg/kg was given 30 minutes before the end of surgery to provide adequate pain relief with minimal risk of respiratory depression but was not sufficient because patients experienced intense pain at recovery.
The most likely explanation for the larger morphine requirements in the large-dose remifentanil group was the development of acute opioid tolerance. Chia et al. (3) suggested that large-dose fentanyl (with halothane) during total abdominal hysterectomy had resulted in higher postoperative pain intensity and greater fentanyl consumption. Guignard et al. (4) established that large dose intraoperative remifentanil administration (with desflurane) during major abdominal surgery was associated with acute opioid tolerance. Our findings differed essentially on two points. First, morphine request was increased only during the titration period; we did not observe any significant difference in morphine consumption by PCA after discharge from the PACU. Second, morphine overconsumption during titration was not associated with an increase in pain scores in patients who had received large-dose remifentanil during anesthesia. In our study, acute opioid tolerance was transient and delayed hyperalgesia did not occur.
Tolerance and hyperalgesia reflect similar phenomena, resulting from opioid-induced central nervous system sensitization. Tolerance corresponds to its pharmacological expression and is manifested by larger morphine requirements to obtain the same pain relief. Hyperalgesia corresponds to its clinical expression and is usually manifested by higher pain scores. However, after an effective and well-conducted treatment of postoperative pain, VAS can no longer reflect hyperalgesia. Thus, in our study, the reduction of the titration delay in the large-dose remifentanil group in the PACU was probably the sole objective finding demonstrating the presence of hyperalgesia.
Ho et al. (22) reported that the magnitude of morphine tolerance was significantly correlated to the duration of opioid infusion. In the study by Guignard et al. (4) the duration of anesthesia (between 288 and 294 minutes) was more than 1 hour longer compared with our study (between 198 and 233 minutes). On the other hand, Cortinez et al. (6) failed to demonstrate any kind of acute opioid tolerance when the duration of anesthesia was shorter than 90 minutes.
The design of our study did not allow a statistical comparison in morphine titration requirement between Parts I and II because patients were not randomly allocated in four groups but rather in two groups within two successive steps. Acute opioid tolerance appeared in Group L receiving the largest doses of remifentanil. In Groups S and Sn receiving the smallest doses of remifentanil, acute opioid tolerance did not occur. The decrease in morphine consumption in Group Sn (with nefopam) compared with Group S could be explained by the analgesic effect of nefopam (23). However, the decrease in morphine consumption in Group Ln (with nefopam) compared with Group L should have been in the same proportion as that found between Groups Sn and S. But acute opioid tolerance did not occur in Group Ln. The lack of statistical difference between Groups Sn and Ln suggests an additional effect of nefopam, which could be interpreted as an inhibition of acute opioid tolerance. Furthermore, a single dose of nefopam eliminated the difference in titration delay.
Several pharmacological mechanisms can underlie opioid-induced pronociceptive effects. Glutamate may be involved in a process of increasing pain sensitivity in the dorsal horn of the spinal cord. NMDA receptors are one of the major subtypes of glutamate receptors. Large-dose opioid may increase glutamate release (1), and µ-opioid receptor stimulation may trigger the activation of NMDA receptors (10,13). In animals (8,9) and in humans (11,12), ketamine reduced morphine requirement and hyperalgesia, suggesting that the NMDA receptors are involved in the generation of acute opioid tolerance. The development of acute opioid tolerance may result not only from activation of the NMDA system but also from the monoaminergic descending inhibitory pathway (24). There is experimental evidence for remifentanil-induced antianalgesia and postinfusion hyperalgesia in volunteers and their differential modulation by ketamine and clonidine (2-adrenergic receptor agonist) (24). Nefopam has both anti-NMDA properties and monoaminergic properties (16). Girard et al. (25) showed that the combination of morphine and nefopam induced antinociceptive synergy in postoperative pain models in rats.
In summary, our findings confirmed that large-dose remifentanil during total IV anesthesia for major abdominal surgery caused postoperative morphine overconsumption. This acute opioid tolerance was transient and occurred in the early postoperative period. We suggest that nefopam could be administrated safely as a coanalgesic treatment during remifentanil and propofol anesthesia to avoid postoperative morphine overconsumption.
|
Footnotes |
|---|
Accepted for publication July 26, 2005.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
